The Nineteenth Byte

I'm doing a bit of background reading on denesting surds with the thought that it might be possible to work in exact values over a quadratic extension of the rationals.

overactor

The results are stunning in any case

are those all with cyclic center definitions?

Peter Taylor

In almost all cases the filename will contain one or two numbers from the Encyclopedia of Triangle Centers. The exceptions are -1, -2, and -3 which I've used for some simple centres which may or may not be in the ETC.

(And may or may not have a sensible geometric interpretation)

overactor

It would be cool to alternate between two definitions following the linus sequence

Peter Taylor

9:08 AM

@overactor Nice idea. Some of them look quite different to the simple alternation. 58,23 develops a pretty good biohazard symbol.

overactor

Pretty scary looking

I figured they would have a different character to it since the sequence is by its very nature non-repeating

I wonder if other sequences do well.

Peter Taylor

Yes, I've got some nice pictures with a Fibonacci tiling from one of the papers on Penrose tilings which I read in August.

The biggest problem with stuff like this is that the search space is so big that you either stumble around looking for cool stuff or you have to learn how to control the features of the fractal.

overactor

9:24 AM

You should make a javascript program that lets people adjust some numbers

And then harness the power of people who have nothing to do with their time.

Peter Taylor

The same thought had crossed my mind.

overactor

How quickly do these images get created?

and in what language is your program written?

Peter Taylor

Varies a bit. 1 second to 10 seconds on a fast computer.

Java. I could port it to JS relatively easily.

overactor

you could also provide parameters for recursion depth and image size

so you can quickly try some things out and make a nice render of something that looks promising

Peter Taylor

Of course, they're already parameters in my code.

Another idea I have, which is more complicated, is that for the centres which are circle centres (incentre, orthocentre, circumcentre, 9-point centre, etc) to draw the circles instead of / as well as the triangles.

overactor

9:36 AM

I can barely imagine what sort of results that will yield

Martin Büttner

@PeterTaylor If you want some JS boilerplate code for 2D graphics in WebGL, let me know ;)

Peter Taylor

I was thinking SVG.

Martin Büttner

I've never used SVG, how responsive is that to interactive input?

Peter Taylor

I'm not planning to write multithreaded JS.

Martin Büttner

I don't expect you to

That doesn't really answer my question though :D

Peter Taylor

9:56 AM

That would be because I don't understand the question.

Martin Büttner

lol okay... I was thinking, if the user varied the parameters smoothly, would it be possible to make an SVG respond in realtime?

(although I think in this case the computation of the centres probably dominates anyway)

Peter Taylor

I don't think there are any parameters which can vary smoothly.

Well, apart from image size, but that's not one I expect people to want to vary smoothly.

Martin Büttner

I don't know... some weights in trilinear coordinate parameterisations?

Peter Taylor

I suppose I could set it up so that you control coordinates in a general 4th or 5th order rational polynomial and then set up the ERC centres which fit that model.

Either way, the rendering is probably going to be too slow for interactive changes.

Martin Büttner

hm, yeah, sounds likely. I think my Mathematica script needed something like 10 seconds for depth 10

overactor

10:03 AM

the position of the original vertices could vary smoothly maybe?

Martin Büttner

but then you get many more problems with infinite coordinates ;)

the equilateral triangle doesn't seem to be as prone to those as some others

overactor

Though I believe Martin found that irregular triangles result in chaos

Martin Büttner

@overactor no, not necessarily... some shapes get bent very organically

I'd really like to try doing a multi-part question, but I can't seem to come up with anything but regex (which would likely have some problems)

New Sandboxed Posts

10:20 AM

0

A: Sandbox for Proposed Challenges

Facing a martian fractal Fractals are awesome. They can be used for a very large number of pratical purposes, from antennas to self-similarity of complex networks. One of the most fantastic uses of fractals tough remains 3D terrain landscaping. With such technique you can create awesome la...

overactor

@MartinBüttner how long does it take before a chatroom is frozen?

1 week?

and is it reversible?

Martin Büttner

mods can reverse it

I don't exactly know how long it takes though

overactor

the marbelous room might be up for freezing soon

Martin Büttner

I'm sure Doorknob will reopen it if we need it

overactor

it seems like everyone's attentionspan sort of ran out

Martin Büttner

10:34 AM

heh, maybe.

overactor

Or maybe there just wasn't much to talk about after @Sparr slowed down with developing his interpreter

If only I had the courage (and time) to start on a javascript interpreter

Martin Büttner

heh, I do have the courage, but currently not the time

I was thinking about doing a JS interpreter, too

overactor

It's the logical next step

Martin Büttner

well it was actually the plan all along, but Sparr wanted to practise his Python ^^

overactor

As good a reason as any to write an interpreter for an esotheric language in Python.

Martin Büttner

10:47 AM

I think I'll make my next simple(ish) string-based challenge rosetta-stone. There has only been one remotely string-based rosetta stone challenge yet. The others have been graphical or mathematical.

hm, I can't find a run-length encoding challenge which allows repetition of more than one character at a time (e.g. cccabababab -> 3c4{ab}). that would be a nice candidate.

@PeterTaylor can I run this idea past your neural PPCG duplicate database? ;)

Peter Taylor

I can't think of any either. What are you going to ask for? Maximum compression according to a scoring system where count string scores len(string) + 1?

Martin Büttner

yeah probably, although I'm not sure how hard that would be when you've got to decide between different groupings. e.g. aaaabcdefabcdef -> 5abcdefabcdef or 4a2{abcdef} ... if that turns out to be too tricky for a code golf, I'd just prescribe a greedy algorithm I think

actually I could just do maximum compression by string length. that would mean that you shouldn't shorten stuff like abab but everything else, I think.

Peter Taylor

I think it could be done by dynamic programming.

Martin Büttner

also I'd like to include nested application: something like aaabbbaaabbb could be shortened to 2{3a3b}

Peter Taylor

Ok, now you're getting tricky.

Martin Büttner

11:01 AM

hehe

too tricky?

probably too tricky for rosetta stone, but maybe not for a plain code golf?

Peter Taylor

11:13 AM

Too tricky for short efficient solutions.

(Possibly for polynomial-time solutions at all).

Martin Büttner

hm yes, so I guess either ask for optimal encoding or for nested encoding.

 xD

Peter Taylor

One advantage of doing SVG over WebGL is that I can draw one triangle (the one which is least likely to have numerical issues) and then add two rotations at +/- 120 degrees.

New Meta Posts

0

Q: Why not renaming the site to cjamgolf?

I think it would be better to rename the site to cjamgolf and I am wondering why it has not been done already. CJam is a programming language, which is optimized for the challenges here. The language reduces the code size by adding implicit assumptions to the language. This is similar to Perl's ...

discussion

overactor

oh boy

Martin Büttner

11:29 AM

wow, he actually signed up on PPCG to say that

overactor

Do you think he'll stick around?

Martin Büttner

probably not. I guess someone should leave a comment, so it feels less like "shut up, we've discussed this before", but I can't be bothered to repeat geobits's argument.

overactor

I'm not much of a golfer so I don't feel like it's my place to comment there

Is devising a challenge to play into a specific programming language's cards frowned upon?

Martin Büttner

hm, depends

in what sense?

overactor

ninja'd

Martin Büttner

11:37 AM

18s, phew ;)

overactor

Say I design a language with a fun feature

for instance

n prints the first n primes in octal base

And then I propose a challenge:

"Write a program that prints the first n primes in octal base"

Martin Büttner

the challenge would be fine

answering it would get you downvotes

because meta.codegolf.stackexchange.com/a/1078/8478 is a valid loophole again

overactor

And I answer my own question in the language I created before the challenge

A good thing if you ask me

What about more subtle things?

Can I devise a challenge that Marbelous would be good at in the hope that I can wheel in a win for Marbelous?

(Not that that would actually result in a win for Marbelous)

Martin Büttner

I don't know... it's hard to tell... you could make someone else post the challenge and so on... there are always ways to cheat your way around the "rules", but in the end, if people notice it, you might get downvotes.

I don't think people like to see that someone posts a challenge for which they already have an unbeatable answer.

overactor

Good point

Peter Taylor

11:54 AM

@MartinBüttner Why feed the troll?

Martin Büttner

@PeterTaylor I don't know. some people might respond to reasonable arguments and just had a wrong impression of how important it is to win a code golf.

overactor

He's likely not a troll, just misguided

overactor

12:16 PM

heh.

Martin Büttner

lol

wolframalpha.com/input/?i=Why%3F&lk=2

overactor

"Because I said so." would have been even better

Martin Büttner

@overactor codegolf.stackexchange.com/questions/5562/… you've been looking for marbelous-solvable challenges, right?

overactor

I'd need to implement 512 bit numbers though

Martin Büttner

lol

I'm sure it's fine if it works for 2^^4

oh it's not

okay, that's a problem

in fact 512 bit numbers aren't even enough

you would just overflow

overactor

12:22 PM

oh right

might as well make it 1024 while I'm at it then

I'd need 128 marbles to call a 1024 bit function

also, for 2^^4 I need more than 16 bit

Martin Büttner

we do have 32 I thought?

overactor

Yeah, but can I just include that or do I need to add that to my char count?

(the implementation)

wolframalpha.com/input/?i=99+bottles+of+beer

pfft

[this one] (wolframalpha.com/input/…) is a beauty.

overactor

1:04 PM

I messed up there..

Martin Büttner

@Doorknob where does the discrepancy between number of questions in Site Stats (sidebar) and the Questions tab come from? does one count locked questions and the other not? they only differ by 13 questions, so it can't be deleted ones.

Geobits

I think it probably is locked. IIRC, locked will not show up in any lists, only with a link.

Martin Büttner

@PeterTaylor What do you think about doing two challenges for advanced RLE? A rosetta stone one which doesn't include nesting and uses a greedy algorithm (aaaaabcdefabcdef would be suboptimally compressed to 5abcdefabcdef), and a code-challenge which includes nesting and should produce the maximum possible compression?

Peter Taylor

1:20 PM

I think the Rosetta Code one would be more interesting with optimality but no nesting.

Martin Büttner

okay. I was just thinking there might be a certain problem complexity that is too much for rosetta stone challenges (because if the first submission is already >300 characters, who can be bothered to port it to 5 more languages). I'll write up a draft later though.

Martin Büttner

2:18 PM

I think Peter may have been right... he's a troll

@MartinBüttner Where is the prove for your statement about J, K and APL? — ceving 29 mins ago

Geobits

And yet the feeding continues... ;)

Martin Büttner

You know how it is... "But someone's wrong on the internet!"

okay... lunchtime is challenge posting time

any objections?

2

A: Sandbox for Proposed Challenges

Word Search Substring code-golf string Given a text (which may contain new lines) and a search string, determine if the text contains the search string. The catch is that you're not just looking for the usual substring, but you're treating the text like a word search puzzle: the substring may a...

Geobits

My only objection is that it's not yet lunchtime here.

Martin Büttner

you aren't posting a challenge though :P

Geobits

Aren't you quite a few hours ahead of me? Seems like a late lunch :P

Martin Büttner

2:23 PM

it is

overactor

The only reasonbably large place where it is a sensible lunchtime at the moment is Nuuk, Greenland.

Martin Büttner

I had a second breakfast at 10.30, so left lunch until a bit later (which also has the benefit of avoiding annoying queues)

Rainbolt

Ugh. Another hoop I have to jump through. Now when I close a bug, I have to record whose fault I think it was.

Martin Büttner

@Rainbolt o.O

a) What purpose does that serve for the working morale of your team?

Peter Taylor

Someone has voted to reopen that incomprehensible GPS question. If it's someone who reads this chat, since your vote obviously means that you don't find it incomprehensible, could you please edit it so that the rest of us can understand what on Earth the OP is asking for?

Martin Büttner

2:26 PM

b) git blame?

Geobits

That sounds like a wonderful teambuilding exercise :D

Rainbolt

Also, I just found out that the emails I have been responding to about "What is the root cause (internal use only)" have been seeing their way to the customer. I am severely pissed off that an internal only comment would make it outside the company.

I have to cool down and eat lunch before I go to my supervisor about it

Martin Büttner

any suggestions for a catchier title than "Word Search Substring"?

"Substring Search Puzzle"?

Geobits

@Martin I don't know that 'substring' needs to be there at all. I think most people know what a word search is, and it's not even really a substring in non-forward-horizontal cases.

Rainbolt

@MartinBüttner It's not that kind of blame. I don't have to identify a person. I have to identify whether it was caused by a poor requirement, a lack of time, human error, etc.

Martin Büttner

2:29 PM

@Rainbolt Ah I see. That makes a bit more sense.

Geobits

Don't you work at Genesis? Obviously the ultimate blame lies in original sin, no?

Rainbolt

I love the Bug Previews because I can easily see the effect it has on my productivity. Two heads studying the direction that needs to be taken to solve the problem

Martin Büttner

@Geobits Just plain old "Word Search Puzzle" then?

or "Solve a Word Search"?

Geobits

To judge a fellow human on this just doesn't seem right.

@Martin I think it's clear enough, but not very catchy (if that's what you want).

Martin Büttner

"Word Search Puzzle" might work well on the HNQ list :P

Rainbolt

2:30 PM

Survival Game - Find Your Word

overactor

@Geobits God didn't blame humans for the original sin?

Martin Büttner

@Rainbolt *own

Geobits

That's true, if it gets to HNQ.

@overactor God's allowed to judge. It's kinda what he does.

Rainbolt

huh?

Martin Büttner

@Rainbolt "Find Your Own Word" (for maximum parallelism with your challenge)

Rainbolt

2:31 PM

Did you know that Lutheran pastors are also allowed to judge, but Methodist and Baptist pastors are not?

overactor

@Geobits Are we allowed to blame Adam and Eve?

Rainbolt

For example, a Lutheran pastor can refuse communion to a specific person during a service. A Baptist pastor cannot

Geobits

@overactor I'm no biblical scholar, but I believe it's something like "judge not your fellow man".

overactor

The word is in another castle?

Geobits

@Rainbolt Tell that to some Baptists I've met :D

Martin Büttner

2:34 PM

@overactor lol

overactor

Psalm 137:9 Happy is the one who seizes your infants and dashes them against the rocks.

Martin Büttner

0

Q: Word Search Puzzle

Given a rectangular text as a word search puzzle and a search string, determine if the text contains the search string. The search string may appear: horizontally, vertically or diagonally forwards or backwards You may write a function or a program and take two strings as input via function a...

code-golf string

lol, 2 views (one of which is mine), 2 upvotes. hmmm.... caching strikes again.

Geobits

@overactor Old v New Testament is a known issue. They definitely flip the switch when it comes to vengeance and retribution.

overactor

Or if you prefer the NET bible: How blessed will be the one who grabs your babies and smashes them on a rock!

Perhaps turn the other cheek meant, turn the babies around when the one cheek is sufficiently smashed?

Rainbolt

Why are brutal baby jokes always so funny? It's disturbing.

Geobits

2:41 PM

Random fact: There's a [seinfeld] tag on ELL, with four questions asked by a guy with Jerry Seinfeld as an avatar.

overactor

That tag hardly seems relevant to the questions he's asking

overactor

2:55 PM

Related Tags: [meaning-in-context] x4
I think it should probably be marked as a synonym

Martin Büttner

lol

New Sandboxed Posts

0

A: Sandbox for Proposed Challenges

Rosetta Stone Challenge: Run-Length Encoding v2.0 rosetta-stone code-golf string compression The goal of a Rosetta Stone Challenge is to write solutions in as many languages as possible. Show off your programming multilingualism! The Challenge We've done run-length encoding before but only co...

Martin Büttner

btw "word search" is such a boring term for those puzzles

overactor

@MartinBüttner You have a better suggestion?

Martin Büttner

the German term translates to "letter salad"

overactor

3:10 PM

is word scramble a thing?

or letter scramble?

Buschtabensalat?

Martin Büttner

s <-> ch

Buchstabensalat

overactor

buchstabensalat is exactly 15 letters

Martin Büttner

lol

overactor

all challenges should have german titles

I bet we could get to 15 letters in one word for every single challenge

Martin Büttner

that's very likely

Lauflängenkodierung

(RLE)

ProgramFOX

3:15 PM

@SohamChowdhury Interesting!

Martin Büttner

3:46 PM

@Geobits can't you do x|y in your if or something?

:D that other answer

Geobits

Yea, but only x|y!=0 I guess. 0/1 isn't truthy/falsey in Java for some inane reason.

Martin Büttner

that's still 4 bytes shorter, though, no?

Geobits

And I have to wrap x|y in parentheses because of precedence ><

Martin Büttner

2 bytes...

Geobits

Oh I know, I'm editing it in now :D

Martin Büttner

4:03 PM

I like how all 3 submissions use a different approach

well no, I guess yours and the second python one are quite similar

Geobits

Pretty similar, yea.

Martin Büttner

Nice Answer badge with 67 views. neat :)

also, yay that'll make for a new digit in my badge summary :D

oh, you're beaten ;)

:D

Java vs C#

IT IS ON!

VisualMelon

aye :P

I didn't realize the other answers until I'd posted mine because the notification for new answers has changed, but I liked that it was 314

Geobits

sigh Time to get to work :D

Rainbolt

I wonder if discussing my company's agile development (or other) practices is a violation of my Oath of Silence

I kind of assumed I wasn't allowed to talk about customer related stuff. I didn't even think about the rest of it.

VisualMelon

4:16 PM

@Geobits please don't :P

I have better things to be doing with my evening ;)

Martin Büttner

@Rainbolt well it's usually intended to cover customer information and details of why your product/service is as awesome as it is... but I think for the specifics you'd have to read your oath of silence.

Peter Taylor

@Geobits Because it has a proper type system, unlike some languages it may be based on?

:P

Geobits

Yea yea. But it's still a pain in the ass sometimes, even when not golfing.

Rainbolt

@MartinBüttner The name is escaping me again

Martin Büttner

@Rainbolt nda

non-disclosure agreement

Peter Taylor

4:25 PM

OTOH it reduces the incidences of a buggy if (a = b) considerably

Martin Büttner

@PeterTaylor a compiler warning would be sufficient for that (for people who heed compiler warnings)

Geobits

Only if your decade-old legacy program doesn't have hundreds of "natural" compiler warnings accumulated over time ><

Martin Büttner

hm, yes, that's sadly true

(another reason to check "Warnings as Errors", so they don't even start to pile up)

Peter Taylor

@MartinBüttner Such people exist?!

Martin Büttner

I'm sure they do. The question is rather whether it's possible to find a team of developers consisting only of such people.

That being said, during my final BSc project, I was working in a team of 6 people working on instrumentation of .NET assemblies for 9 months and our code was warning-free until the end

(Warnings as Errors ftw!)

Rainbolt

4:33 PM

We have thousands of warnings that were created when we 1) Switched to a new version of Visual Studio, 2) Switched to a new version of .Net, or 3) Created by people who don't care

The first two are actually legitimate excuses IMO

VisualMelon

unless it's one of those .NET warnings regarding breaking changes

Martin Büttner

Well the first one might be, but ideally a framework port shouldn't be done half-assed either ;) (going from no warnings to thousands of warnings for the sake of upgrading the framework... meh)

VisualMelon

my C++ is always full of warnings because I don't bother explicitly casting stuff because I don't care if it's going to be truncated or not, I might also not know how to iterate through vectors properly

mniip

6:21 PM

Yo.

Martin Büttner

7:11 PM

No.

Rainbolt

So?

Yohoho! Puzzle Pirates

Every single one of the challenges in that game could have bots written for them. They boil down to search trees, combinatorics, and probability. Perhaps I should start a series?

Martin Büttner

7:30 PM

Please do!

(I don't know puzzle pirates, but yay for series)

Rainbolt

They wouldn't be unique problems exactly. But I'll give it a shot

user2179021

hi

overactor

7:52 PM

evenin'

mniip

bacon

Flonk

8:22 PM

I don't understand the obsession with bacon america has

I mean bacon is cool, but meh

There's better

Geobits

@Visual Ok, I got mine down to 211. Your turn ;)

If I felt using char[] instead of String wasn't borderline, I could save quite a few more.

Martin Büttner

8:45 PM

@Flonk Blasphemy!

(I'm not American)

Rainbolt

@Flonk is not longer allowed to take communion.

Flonk

._. I'm sorry.

Rainbolt

@Flonk Confess your sinbacon!

Flonk

9:04 PM

It's just that there are so many great meat products! Bacon seems really.. mediocre

(I love myself some chorizo though)

Rainbolt

@Flonk Did you know that bacon became the icon of a new way to describe food in Japan sometime within the last decade?

There was sweet, salty, and sour. Now there is "meaty" (not the best translation)

And bacon is the ultimate meaty food.

Meatier than steak.

Flonk

@Rainbolt I've heard about Umami but I didn't know it was connected to bacon

That is kind of awesome I guess!

Rainbolt

Yes! That's it

I guess it's older than a decade

Wait

I think that's the right one. I'm not sure.

I heard about all of this from a friend who triple majored in Japanese, Radio/Television/Film, and Mathematics

thewateringmouth.com/what-is-umami-why-does-bacon-taste-so-good

I missed bitter. I didn't really think something could be bitter and sour at the same time.

Can anyone identify a food that is both sour and bitter?

New challenge posted: Output the name of a food that is both sour and bitter. code-golf

Flonk

Hmm glutamate has some negative connotations (at least over here).. They put it in fast food so it tastes better

Which explains a lot I guess

Bitter lemon is both bitter and sour, I'd say

Rainbolt

One would think. I've never had one

I wonder if it's possible to create a dish that has all five!

If the powers of each flavor were matched precisely, would it taste like all five, or would it taste like nothing at all?

2

Flonk

9:20 PM

Haha

My guess is that it would just be super intense and you wouldn't actually taste much

There should be food-golf

Rainbolt

We've created a new flavor: intense!

Powerthirst may have discovered it first

New Sandboxed Posts

0

A: Sandbox for Proposed Challenges

Can you lose from this chess position? code-challenge chess Introduction You are an arbiter who supervises a very large number of chess tournaments. Thus, you frequently rule on whether a player who runs out of time may claim a draw based on Article 6.9 of the FIDE Laws of Chess: [...] If ...

Flonk

@Rainbolt I'd buy it.

Peter Taylor

Bitter + sour is an unusual combination because many bitter foods are alkaline, whereas sour foods are acidic.

But balancing the flavours is a standard part of food design.

Rainbolt

@PeterTaylor I thought perhaps a food might exist where they were balanced but not mixed well, such that each flavor hits different parts of the tongue and you get bitter AND sour at the same time.

Flonk

9:30 PM

@PeterTaylor Oh, I did not know that!

Peter Taylor

One of my favourite simple snacks is green olives, lemon juice, small bits of toasted almond, and tarragon. It doesn't have much umami, but the other four basics are represented. The olives give you bitterness and saltiness, the lemon juice gives sourness, the almond and tarragon are slightly sweet.

Rainbolt

Nuts and olives are both considered umami

Seems like the Internet thinks everything is umami.

Peter Taylor

On another note, have people seen that the IOCCC is open?

Flonk

I wasn't even aware that the IOCCC still exists!

Rainbolt

I wonder why spicy isn't a basic taste. Is it a combination of other tastes?

New Meta Posts

9:40 PM

0

Q: Proposing new winning criterion tag: Voter-judged

I'm stepping into a hornet's nest here. Popularity contests have been controversial. Some detractors believe that non-objective winning criteria are simply unacceptable. Others are OK with subjectivity in theory, but dislike that voters upvote answers based on coolness rather than programming me...

discussion tags winning-criteria popularity-contest

xnor

I'm here to discuss the surely-controversial proposal I just posted.

Rainbolt

Ah, I found it: The tongue can also feel other sensations not generally included in the basic tastes. These are largely detected by the somatosensory system. ~ Wikipedia

Peter Taylor

Spiciness is complicated.

Capsaicin triggers the same sensors as physical heat.

Mustard is mainly sensed in the nasal passages.

Rainbolt

@xnor Your proposal must be at least as good as popularity contests.

@xnor I won't speak to the validity of either. Just that I think your's is strictly better.

xnor

@Rainbolt thanks

@Rainbolt Do you think pop cons should not exist in general?

Rainbolt

9:47 PM

Well I wanted to avoid talking about that, but I think they should exist.

xnor

Fair enough. I have no particular affinity for pop cons, but I hope that it will help to separate out the question of objectivity from that of freeform voting.

Rainbolt

If the criterion upon which you are asking users to vote is objective, then you could easily apply it yourself. So I assume you are going to apply a subjective criterion?

(I will inevitably mess up the word criteria/criterion and not catch it before the edit timer runs out. Apologies in advance.)

xnor

Yes, though it's open to what degree it's subjective. On the more objective side would be "Is all the code relevant?" for xode-structure challenges like palindromes where comments and unused variables make it silly. On the more subjective side could be the code golf animation gif challenge if specific aspects (clarity, intelligence of matching, syntax highlighting) were mentioned as the criteria.

Rainbolt

You might clarify in your proposal. Everywhere that you wrote the word "criteria", I'd want to know whether you meant objective or subjective.

And how many criteria there could be. We generally close questions that have conflicting criteria.

xnor

thanks, editted

I'd expect some questions would have multiple criteria, saying how much to emphasize each one

Rainbolt

9:57 PM

And you might speak about down votes. If someone doesn't meet the criteria, do you down vote?

xnor

tradeoffs are something that humans are good at judging, whereas point-based systems tend to have corner cases where you just want to maximize one or the other

yes, i should add that

oh, i have to run, i'll be back in ~30 min

Rainbolt

@xnor Run farther than you did last time!

That's my golden rule and it works wonders.

Flonk

How much rep is an accepted answer?

Martin Büttner

@Flonk for whom?

Flonk

The one who posted the answer

Rainbolt

10:00 PM

If you accept:

someone else's answer: You gain +2 reputation and the author of the accepted answer earns +15 reputation.
your own answer: No reputation is awarded, and the answer does not float to the top of the list. You must wait 48 hours to accept your own answer.
a community-wiki answer: No reputation is awarded.

source

Flonk

Thats not a lot... Why is keeping winning criteria objective so important then, when the lion's share of rep is awarded by the community anyways?

Martin Büttner

I know I'm a bit late, but: spiciness is basically just pain.

Rainbolt

I thought the ability to win fairly was just an essential part of a challenge.

Pain isn't a basic taste?!

DigitalTrauma

Hey @ma

Martin Büttner

@Rainbolt I like "savoury" as a translation

DigitalTrauma

10:08 PM

For the wordsearch question

you mention return of truthy/falsey

and link to Peter's answer to the meta question

Does that mean you specifically want this meaning of truthy/falsey

trying to figure out if you're ok with omitting the -q to the grep in my answer

and thus return of empty string means FALSE

non-empty means TRUE

Flonk

@Rainbolt Well imo inventing languages to win code-golf challenges isn't particularly fair either.

CJam etc

I just think this whole discussion about objective criteria is a little silly; PPCG is a place that emphasizes the fun parts about programming, and that's why people are here for. From what I gather there are people who even dislike popularity contests, which generally get loads of upvotes (i.e. the community wants to see them)

Martin Büttner

@DigitalTrauma oh, that ping was directed to me... needs three characters ;)

@DigitalTrauma yes, I do want that. i.e. whatever's truthy/falsy in the context of your language (and works in conditionals as such)

DigitalTrauma

@MartinBüttner ok thanks

xnor

10:27 PM

@Flonk I agree that winning criteria can be strange in practice. For code golfs, the criterion is sometimes effectively "Whoever writes the Golfscript/Cjam answer wins!"

@Flonk That said, I think there is something to be said for exact conditions as a goal to work towards, even if you know you won't realistically win.

Flonk

10:52 PM

@xnor I see what you mean, although I'm not sure that outweighs the negatives.. These objective criteria are usually pretty awkward - there is an obvious flaw in the criterion for code-golf (because CJam). The community usually does a good job at upvoting the answers that were 'golfed' the most, so why not let them decide instead of relying on bytecount?

I'd suspect the question asker to do a good job at selecting a winner aswell, so you'd actually see interesting answers as the accepted ones rather than the guy who used the language that has a 1-char-function for reading a line from stdin and pushing it onto the stack

(Even if he doesn't thats just 15 rep going to waste)

Peter Taylor mentioned the IOCCC - those guys don't have an objective winning criterion and the winners' entries are usually pretty cool

So I don't think having "try to write the shortest code and I'll select a cool entry as the winner" would make the quality of answers (or questions) decrease

xnor

@Flonk I'm with you except for "The community usually does a good job at upvoting the answers that were 'golfed' the most". Often, whatever is posted first gets the most upvotes, and shorter languages still get more votes regardless of relative golfing. I like the idea of the question asker awarding the answer to whatever they like. They are certainly in a position to judge quality. However, maybe they can't be qualified to judge different languages.

Flonk

I suppose you are right about that

xnor

11:07 PM

Regarding acceptance, I'm entertained that the accepted winning answer for codegolf.stackexchange.com/questions/37851/… was a direct port of my answer into a more compact language.

No complaints, I got crazy upvotes and the populist badge.

But it's funny that the competition was to write a version of it in something like Pyth or cjam.

Especially in compression Python to Pyth, which was made to make Python answers short.

Flonk

Well, that's because the winning criterion is objective but not fair

I'll skim through meta tomorrow and see how often this has already been discussed (lots, I guess), but for now I'll go to bed since it's already 1am.

Bye!

xnor

good night!

Transcript for